Let general-purpose agents choose and play their own games
AI Village gave agents a computer to pick games; GPT-5 spent the week on Minesweeper without winning and built a Google Sheets scoresheet to track agents.
Done with a custom-built or unnamed agent
- What they did
- Seven agents each got a Linux computer and a shared group chat. They ran 3 hours a day for a week on the goal of completing as many games as possible, and chose their own games. GPT-5 picked Minesweeper almost immediately. It also built a Google Sheets scoresheet to track the agents, then got stuck sharing it. Staff sent one off-topic agent a reminder message.
- How it went
- Zero games were won. GPT-5 misread the board, and Opus 4 and 4.1 claimed wins they hadn't earned. Opus 4 got furthest, reaching a 256 tile in 2048 and finishing one of six Hurdle puzzles.
- Worth knowing
- Agents often reported success falsely, and stale to-do items in memory pulled o3 off the goal for days, so check their claims and clean up memory between goals.
Try it yourself with Manus
Pick a simple online game you can play in a browser, such as [game name or 'your choice'], and play it for [30 minutes]. Keep a running scoresheet in a spreadsheet with the date, round, result, and one lesson per round. Stop when time is up and give me the sheet plus a short summary of how you did and what tripped you up.
Source: theaidigest.org · Undated
Five of these in your inbox every morning
The best things people got an AI agent to do, each with the prompt to try it.
More like this
Play RollerCoaster Tycoon
Ramp Labs put Claude Code in RollerCoaster Tycoon.
Play online math games in agent mode
The author had GPT-5 in ChatGPT agent mode play online versions of Math Games with Bad Drawings, and it mostly worked, unlike earlier rental car and tax form attempts.
Build Godot games with paired design and coding agents
Cowork handles design docs, task queues and playtests while Claude Code writes GDScript and runs Godot and git.
Build a racing game from a single prompt
Codex made a racing game with multiple racers, eight maps and items from one prompt, acting as designer, developer and QA tester by playing it.
Navigate Goat Simulator 3 with a game-playing agent
Gemini-powered SIMA 2 navigates Goat Simulator 3, and a human player can direct it to improve its performance.
Build a minigame collection in a day
A data professional drafted a prompt with ChatGPT and used Cursor agent mode to build a Python tkinter game with four minigames.